concept
active
concept:task-generalizationtask generalization
The ability to generalize across tasks; lacking in latent methods.
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Identifies key limitations of latent methods.
Concepts (2)
concept
- Generalisationrelated_toAbility to respond appropriately to novel situations based on past regularities; fundamental to learning and intelligence.
- Generalizationrelated_toAbility to apply learned solutions to novel circumstances.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Measuring AUROC of a probe trained on one task when evaluated on another task to assess universality.
- Abstracting from specific memories (e.g., specific leaves) to general lessons (food).
- Generalization from 2-digit to 3-4 digit arithmetic; limited by mismatch dr.
- The ability of probes trained on one dataset to transfer accurately to topically and structurally different datasets
- Novel task asking which of 10 sentences received injection, cycling injection through all positions to average out positional bias
- Models learning goals that generalize undesirably outside training distribution; alignment faking is a challenging instance
- Core assertion about the value of learning.
- Evaluation of learned circuits on grids 4x larger with 4x more steps than training conditions